medium_cross.en

This model is a fine-tuned version of crossdelenna/medium_cross.en on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3034
• Wer: 15.1384

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 22
• eval_batch_size: 22
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10
• training_steps: 1051
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.664	1.2411	350	0.3998	18.2094
0.4625	2.4823	700	0.3244	16.0633
0.3703	3.7234	1050	0.3034	15.1384

Framework versions

• Transformers 4.49.0.dev0
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0